CRUM: Checkpoint-Restart Support for CUDA’s Unified Memory

Rohan Garg, Apoore Mohan, Michael Sullivan, Gene Cooperman
Northeastern University, Boston, USA
arXiv:1808.00117 [cs.DC], (1 Aug 2018)


   title={CRUM: Checkpoint-Restart Support for CUDA’s Unified Memory},

   author={Garg, Rohan and Mohan, Apoore and Sullivan, Michael and Cooperman, Gene},






Download Download (PDF)   View View   Source Source   



Unified Virtual Memory (UVM) was recently introduced on recent NVIDIA GPUs. Through software and hardware support, UVM provides a coherent shared memory across the entire heterogeneous node, migrating data as appropriate. The older CUDA programming style is akin to older large-memory UNIX applications which used to directly load and unload memory segments. Newer CUDA programs have started taking advantage of UVM for the same reasons of superior programmability that UNIX applications long ago switched to assuming the presence of virtual memory. Therefore, checkpointing of UVM will become increasingly important, especially as NVIDIA CUDA continues to gain wider popularity: 87 of the top 500 supercomputers in the latest listings are GPU-accelerated, with a current trend of ten additional GPU-based supercomputers each year. A new scalable checkpointing mechanism, CRUM (Checkpoint-Restart for Unified Memory), is demonstrated for hybrid CUDA/MPI computations across multiple computer nodes. CRUM supports a fast, forked checkpointing, which mostly overlaps the CUDA computation with storage of the checkpoint image in stable storage. The runtime overhead of using CRUM is 6% on average, and the time for forked checkpointing is seen to be a factor of up to 40 times less than traditional, synchronous checkpointing.

Recent source codes

* * *

* * *

HGPU group © 2010-2020 hgpu.org

All rights belong to the respective authors

Contact us: